A Novel Clustering Algorithm Based on Gravity and Cluster Merging

摘要：

Fuzzy C-means (FCM) clustering algorithm is commonly used in data mining tasks. It has the advantage of producing good modeling results in many cases. However, it is sensitive to outliers and the initial cluster centers. In addition, it could not get the accurate cluster number during the algorithm. To overcome the above problems, a novel FCM algorithm based on gravity and cluster merging was presented in this paper. By using gravity in this algorithm, the influence of outliers was minimized and the initial cluster centers were selected. And by using cluster merging, an appropriate number of clustering could be specified. The experimental evaluation shows that the modified method can effectively improve the clustering performance.

关键词： Fuzzy C-means Algorithm Gravity Cluster Merge

作者: Jiang Zhong Longhai Liu Zhiguo Li

作者单位: College of Computer Science Chongqing University Chongqing 400044 China Shanghai Baosight Software Corporation,Chongqing 400039 China

会议类型: 国际会议

会议名称: 6th International Conference on Advanced Data Mining and Applications(第六届先进数据挖掘及应用国际会议 ADMA 2010)

会议地点: 重庆

会议语种:英文

页码: 302-309

在线出版日期: 2010-11-19（万方平台首次上网日期，不代表论文的发表时间）